[cvsnt] process unicode file incorrectly

[Bo Berglund]

When I have a file that has not changed contents but has a changed 
timestamp and is thus flagged as modified in WinCvs, what I usually 
do is cvs status and cvs sees that there are in fact no content 
changes and updates the Entries file to reflect the file's changed 
timestamp. Then the modified state in WinCvs is removed.

I think that the OP is referring to this behaviour and that it 
does not work for unicode files.

Best regards,

Bo Berglund


-----Original Message-----
From: cvsnt-bounces at cvsnt.org [mailto:cvsnt-bounces at cvsnt.org] On Behalf
Of Tony Hoyle
Sent: den 30 augusti 2006 12:34
To: cvsnt at cvsnt.org
Subject: Re: [cvsnt] process unicode file incorrectly

Winston Feng wrote:
> CVSNT version: 2.5.03 (Scorpio) Build 2382
> 
> bug 1:
> A ucs-2 file a.txt:
> 1. cvs add -ku a.txt
> 2. cvs ci -m "" a.txt
> 3. touch the a.txt
> 4. cvs st a.txt, get:
>   Status: Locally Modified

Looks fine. What is wrong?

> bug 2:
> if the a.txt get conflict, cvs will corrupt the file.
> See the attachment.
>
No attachments on the list.  I have tried this and it works OK.  Are you

confusing conflict markers for corruption perhaps?

Tony

_______________________________________________
cvsnt mailing list
cvsnt at cvsnt.org
http://www.cvsnt.org/cgi-bin/mailman/listinfo/cvsnt